Lines Matching refs:FIFO

169 	 * Since this status did not consume a FIFO, we have to
171 	 * to this transaction.  There are two states that a FIFO still
174 	 * 1) Configured and draining to the host, with a FIFO handler.
177 	 * Case 1 can be detected by noticing a non-zero FIFO active
179 	 * the FIFO to complete the SCB.
182 	 * pointers for this same context in the other FIFO.  So, if
308 	 * The FIFO use count field is shared with the
583  * Allocate a FIFO for a non-packetized transaction.
585  * can allocate a FIFO for a non-packetized transaction.
589 	 * Do whatever work is required to free a FIFO.
874 	 * Command retry.  Free our current FIFO and
875 	 * re-allocate a FIFO so transfer state is
1233  * Complete the current FIFO's SCB if data for this same
1234  * SCB is not transferring in the other FIFO.
1379  * code do the rest.  We also reset/reallocate the FIFO to make
1471 	 * Can this FIFO have access to the S/G cache yet?
1482 	 * Should the other FIFO get the S/G cache first?  If
1484 	 * any FIFO, it is important that we service a FIFO
1486 	 * that a FIFO will be freed to handle snapshot requests for
1487 	 * any FIFO that is still on the bus.  Chips with RTI do not
1494 		 * this FIFO.
1499 		 * Switch to the other FIFO.  Non-RTI chips
1507 		 * If the other FIFO needs loading, then it
1510 		 * the original FIFO mode and we test this above).
1512 		 * FIFO not currently on the bus first.
1543 	 * request in the other FIFO.
1555 	 * first segment in the S/G FIFO.  Wait until it is
1657 	 * Ensure that any FIFO contents are cleared out and the
1658 	 * FIFO free'd prior to starting the BITBUCKET.  BITBUCKET
1659 	 * doesn't discard data already in the FIFO.
1718 	 * some overrun bytes into the data FIFO and acked them
1721 	 * transfer and then test to see if the data FIFO is
1858 		 * we have already prefetched data into our FIFO (LQ/Data
1864 		 * a continuation of a transfer already setup in our FIFO
1879 		 * Switch to the active FIFO after clearing the snapshot
1880 		 * savepointer in the current FIFO.  We do this so that
1882 		 * FIFO.  This status is the only way we can detect if we
1910  * The first is a snapshot save pointers where the current FIFO is not
1918  * The second case is a save pointers on an active FIFO which occurs
1987  * ack bytes into the FIFO, and may even ack the last overrun packet
1988  * into the FIFO.   If the FIFO should become non-empty, we are in
2005 	 * can be sure it pertains to this FIFO.  SAVEPTRS
2008 	 * shadow is not valid, keep running this FIFO until we
2026  * Either a SAVEPTRS interrupt condition is pending for this FIFO
2027  * or we have a pending NONPACKREQ for this FIFO.  We differentiate
2046 	 * Keep a handler around for this FIFO until it drains
2061  * LAST_SEG_DONE status has been seen in the current FIFO.
2095 	 * the check to see if another FIFO is active because
2097 	 * the FIFO anyway since it costs us only one extra
2147 	 * a NONPACKREQ phase change have occurred and the FIFO is
2173 	 * data channel use will have a FIFO reference count.  It
2195 	 * in a FIFO or an outgoing LQ.  Only treat it as an I_T only